Qwen3 235B A22B Thinking 2507 vs Qwen3.5 122B A10B
Qwen3 235B A22B Thinking 2507 and Qwen3.5 122B A10B match in llmrun Score, with values of 108.1 versus 109.4; neither fits 24 GB.
Model A · Alibaba
Qwen3 235B A22B Thinking 2507235.1B · 22.2B active · 256K context · Apache 2.0 · Jul 2025
Model B · Alibaba
Qwen3.5 122B A10B125.1B · 12.7B active · 256K context · Apache 2.0 · Feb 2026
Choose Qwen3 235B A22B Thinking 2507 if you want…
Stronger at science
107.7 vs 96.0
Also does chat
Qwen3.5 122B A10B doesn't list it
Choose Qwen3.5 122B A10B if you want…
Stronger at reasoning
112.4 vs 105.9
Ahead on the benchmarks
5 of 6 shared boards
Fits a Mac 64 GB
IQ3_M · 56.7 GB; Qwen3 235B A22B Thinking 2507 needs 72.4 GB at its smallest quant
Lighter on memory
75.4 vs 141.6 GB at Q4_K_M
Also does vision
Qwen3 235B A22B Thinking 2507 doesn't list it
Newer release
Feb 2026 vs Jul 2025
On your hardware
Each model at its default quant (Q4_K_M), a smaller one when it must. Estimates, ±20%.
8 GB
RTX 4060 · 8 GB tier
Qwen3 235B A22B Thinking 2507
Doesn't fit
smallest quant needs 72.4 GB
Qwen3.5 122B A10B
Doesn't fit
smallest quant needs 34.9 GB
12 GB
RTX 3060 12GB · 12 GB tier
Qwen3 235B A22B Thinking 2507
Doesn't fit
smallest quant needs 72.4 GB
Qwen3.5 122B A10B
Doesn't fit
smallest quant needs 34.9 GB
16 GB
RTX 5060 Ti 16GB · 16 GB tier
Qwen3 235B A22B Thinking 2507
Doesn't fit
smallest quant needs 72.4 GB
Qwen3.5 122B A10B
Doesn't fit
smallest quant needs 34.9 GB
24 GBreference
RTX 4090 · 24 GB tier
Qwen3 235B A22B Thinking 2507
Doesn't fit
smallest quant needs 72.4 GB
Qwen3.5 122B A10B
Doesn't fit
smallest quant needs 34.9 GB
32 GB
RTX 5090 · 32 GB tier
Qwen3 235B A22B Thinking 2507
Doesn't fit
smallest quant needs 72.4 GB
Qwen3.5 122B A10B
Doesn't fit
smallest quant needs 34.9 GB
48 GB
RTX A6000 · 48 GB tier
Qwen3 235B A22B Thinking 2507
Doesn't fit
smallest quant needs 72.4 GB
Qwen3.5 122B A10B
IQ2_M · ~88 tok/s
42.7 GB · up to 256Kfits
Mac 32 GB
Mac Mini M4 · Macs
Qwen3 235B A22B Thinking 2507
Doesn't fit
smallest quant needs 72.4 GB
Qwen3.5 122B A10B
Doesn't fit
smallest quant needs 34.9 GB
Mac 64 GB
Mac Studio M4 Max · Macs
Qwen3 235B A22B Thinking 2507
Doesn't fit
smallest quant needs 72.4 GB
Qwen3.5 122B A10B
IQ3_M · ~41 tok/s
56.7 GB · up to 256Kfits
Mac 128 GB
Mac Studio M4 Max · Macs
Qwen3 235B A22B Thinking 2507
Q3_K_L · ~21 tok/s
122.4 GB · up to 32K
Qwen3.5 122B A10B
Q4_K_M · ~34 tok/s
75.5 GB · up to 256Kmore headroom
Reference GPU: RTX 4090. Tinted cell = the better pick on that machine (fits, more headroom, or at least 10% faster).
Overall
Qwen3 235B A22B Thinking 2507
Qwen3.5 122B A10B
Coding
Qwen3 235B A22B Thinking 2507
Qwen3.5 122B A10B
no data
Science
Qwen3 235B A22B Thinking 2507
Qwen3.5 122B A10B
Reasoning
Qwen3 235B A22B Thinking 2507
Qwen3.5 122B A10B
Qwen3 235B A22B Thinking 2507 · #31 of 92 open models · #15 of 71 that fit ≤192 GB
Qwen3.5 122B A10B · #25 of 92 open models · #6 of 54 that fit ≤96 GB
Benchmarks
6 shared boards: Qwen3 235B A22B Thinking 2507 leads 1, Qwen3.5 122B A10B leads 5 · Benchmarks imported Oct 2026
Qwen3 235B A22B Thinking 2507
1400.0#48/186 open
ArenaLMArena · Oct 2026
Qwen3.5 122B A10B
1416.3#37/186 open
ArenaLMArena · Oct 2026
Qwen3 235B A22B Thinking 2507
84.5%#11/144 open
IndependentTIGER-Lab
Qwen3.5 122B A10B
86.7%#4/144 open
IndependentTIGER-Lab
Qwen3 235B A22B Thinking 2507
42.4%#26/65 open
Lab-reportedEpoch AI · Jul 2025
Qwen3.5 122B A10B
35.6%#43/65 open
Lab-reportedEpoch AI · Feb 2026
Qwen3 235B A22B Thinking 2507
0.0%#62/65 open
Lab-reportedEpoch AI · Jul 2025
Qwen3.5 122B A10B
0.9%#35/65 open
Lab-reportedEpoch AI · Feb 2026
Qwen3 235B A22B Thinking 2507
80.3%#22/68 open
Lab-reportedEpoch AI · Jul 2025
Qwen3.5 122B A10B
84.3%#17/68 open
Lab-reportedEpoch AI · Feb 2026
Qwen3 235B A22B Thinking 2507
29.3%#23/54 open
Lab-reportedEpoch AI · Jul 2025
Qwen3.5 122B A10B
32.2%#18/54 open
Lab-reportedEpoch AI · Feb 2026
9 more boards (other shared and single-model boards)
Qwen3 235B A22B Thinking 2507
9.0%#13/18 open
IndependentEpoch AI · Jul 2025
Qwen3.5 122B A10B
17.0%#12/18 open
IndependentEpoch AI · Feb 2026
Only Qwen3 235B A22B Thinking 2507 (7)
- GPQA Diamond80.0%
- AIME 2024/202586.7%
- SimpleQA40.4%
- WeirdML41.0%
- Chess Puzzles12.0%
- Fiction.LiveBench75.0%
- Vending-Bench 2-11.3
Only Qwen3.5 122B A10B (1)
- LMArena WebDev1360.4
VRAM by quantization
Memory to load the model and 2K context; below it, the total at 32K and 128K context. Lighter side in bold.
Q2_K
Qwen3 235B A22B Thinking 2507
100 GB85.7 GB file
32K 103 GB · 128K 113 GB
Qwen3.5 122B A10B
53.5 GB46.4 GB file
32K 53.8 GB · 128K 54.7 GB
Q3_K_S
Qwen3 235B A22B Thinking 2507
103 GB101.4 GB file
32K 106 GB · 128K 116 GB
Qwen3.5 122B A10B
55.1 GB52.5 GB file
32K 55.3 GB · 128K 56.2 GB
Q3_K_M
Qwen3 235B A22B Thinking 2507
115 GB112.4 GB file
32K 118 GB · 128K 128 GB
Qwen3.5 122B A10B
61.4 GB56.4 GB file
32K 61.6 GB · 128K 62.5 GB
Q4_0
Qwen3 235B A22B Thinking 2507
118 GB133.1 GB file
32K 121 GB · 128K 130 GB
Qwen3.5 122B A10B
62.9 GB72.5 GB file
32K 63.2 GB · 128K 64.1 GB
Q4_K_M
Qwen3 235B A22B Thinking 2507
142 GB142.2 GB file
32K 145 GB · 128K 154 GB
Qwen3.5 122B A10B
75.4 GB76.5 GB file
32K 75.7 GB · 128K 76.6 GB
Q5_K_M
Qwen3 235B A22B Thinking 2507
168 GB166.8 GB file
32K 171 GB · 128K 180 GB
Qwen3.5 122B A10B
89.5 GB91.5 GB file
32K 89.7 GB · 128K 90.6 GB
Q6_K
Qwen3 235B A22B Thinking 2507
194 GB193.0 GB file
32K 197 GB · 128K 207 GB
Qwen3.5 122B A10B
104 GB101.0 GB file
32K 104 GB · 128K 105 GB
Q8_0
Qwen3 235B A22B Thinking 2507
236 GB249.9 GB file
32K 239 GB · 128K 248 GB
Qwen3.5 122B A10B
125 GB129.9 GB file
32K 126 GB · 128K 127 GB
18 more quants
IQ2_XXS
Qwen3 235B A22B Thinking 2507
not listed
Qwen3.5 122B A10B
34.8 GB36.6 GB file
32K 35.0 GB · 128K 35.9 GB
IQ2_XS
Qwen3 235B A22B Thinking 2507
71.0 GB65.6 GB file
32K 74.0 GB · 128K 83.5 GB
Qwen3.5 122B A10B
37.9 GB40.0 GB file
32K 38.1 GB · 128K 39.0 GB
IQ2_S
Qwen3 235B A22B Thinking 2507
74.0 GB66.1 GB file
32K 76.9 GB · 128K 86.4 GB
Qwen3.5 122B A10B
39.5 GB40.6 GB file
32K 39.7 GB · 128K 40.6 GB
IQ2_M
Qwen3 235B A22B Thinking 2507
79.8 GB78.6 GB file
32K 82.8 GB · 128K 92.3 GB
Qwen3.5 122B A10B
42.6 GB39.1 GB file
32K 42.8 GB · 128K 43.7 GB
IQ3_XXS
Qwen3 235B A22B Thinking 2507
91.6 GB93.1 GB file
32K 94.6 GB · 128K 104 GB
Qwen3.5 122B A10B
48.9 GB44.7 GB file
32K 49.1 GB · 128K 50.0 GB
Q2_K_S
Qwen3 235B A22B Thinking 2507
not listed
Qwen3.5 122B A10B
50.4 GB41.7 GB file
32K 50.6 GB · 128K 51.5 GB
IQ3_XS
Qwen3 235B A22B Thinking 2507
97.5 GB96.9 GB file
32K 100 GB · 128K 110 GB
Qwen3.5 122B A10B
52.0 GB59.2 GB file
32K 52.2 GB · 128K 53.1 GB
IQ3_S
Qwen3 235B A22B Thinking 2507
not listed
Qwen3.5 122B A10B
53.5 GB46.6 GB file
32K 53.8 GB · 128K 54.7 GB
IQ3_M
Qwen3 235B A22B Thinking 2507
106 GB107.3 GB file
32K 109 GB · 128K 119 GB
Qwen3.5 122B A10B
56.7 GB61.6 GB file
32K 56.9 GB · 128K 57.8 GB
Q3_K_L
Qwen3 235B A22B Thinking 2507
121 GB111.2 GB file
32K 124 GB · 128K 133 GB
Qwen3.5 122B A10B
64.5 GB61.6 GB file
32K 64.7 GB · 128K 65.6 GB
IQ4_XS
Qwen3 235B A22B Thinking 2507
127 GB125.5 GB file
32K 130 GB · 128K 139 GB
Qwen3.5 122B A10B
67.6 GB60.2 GB file
32K 67.8 GB · 128K 68.7 GB
IQ4_NL
Qwen3 235B A22B Thinking 2507
133 GB133.1 GB file
32K 136 GB · 128K 145 GB
Qwen3.5 122B A10B
70.7 GB61.4 GB file
32K 71.0 GB · 128K 71.9 GB
Q4_1
Qwen3 235B A22B Thinking 2507
133 GB147.2 GB file
32K 136 GB · 128K 145 GB
Qwen3.5 122B A10B
70.7 GB79.7 GB file
32K 71.0 GB · 128K 71.9 GB
Q4_K_S
Qwen3 235B A22B Thinking 2507
133 GB133.7 GB file
32K 136 GB · 128K 145 GB
Qwen3.5 122B A10B
70.7 GB71.7 GB file
32K 71.0 GB · 128K 71.9 GB
Q4_K_L
Qwen3 235B A22B Thinking 2507
not listed
Qwen3.5 122B A10B
77.0 GB78.2 GB file
32K 77.2 GB · 128K 78.1 GB
Q5_K_S
Qwen3 235B A22B Thinking 2507
162 GB161.9 GB file
32K 165 GB · 128K 175 GB
Qwen3.5 122B A10B
86.4 GB86.4 GB file
32K 86.6 GB · 128K 87.5 GB
Q5_K_L
Qwen3 235B A22B Thinking 2507
not listed
Qwen3.5 122B A10B
91.1 GB90.9 GB file
32K 91.3 GB · 128K 92.2 GB
BF16
Qwen3 235B A22B Thinking 2507
471 GB470.3 GB file
32K 474 GB · 128K 483 GB
Qwen3.5 122B A10B
251 GB244.3 GB file
32K 251 GB · 128K 252 GB
Qwen3 235B A22B Thinking 2507
Ollama · Q4_K_M
ollama run hf.co/unsloth/Qwen3-235B-A22B-Thinking-2507-GGUF:Q4_K_MQwen3.5 122B A10B
Ollama · Q4_K_M
ollama run hf.co/unsloth/Qwen3.5-122B-A10B-GGUF:Q4_K_M| Spec | Qwen3 235B A22B Thinking 2507 | Qwen3.5 122B A10B |
|---|---|---|
| Publisher | Alibaba | Alibaba |
| Parameters | 235.1B | 125.1B |
| Active parameters | 22B per token | 13B per token |
| Architecture | Qwen3MoeForCausalLM | Qwen3_5MoeForConditionalGeneration |
| Context | 256K tokens | 256K tokens |
| Modalities | Chat, Function-calling | Vision, Function-calling |
| License | Apache 2.0 | Apache 2.0 |
| Released | Jul 2025 | Feb 2026 |
Frequently Asked Questions
Which is better, Qwen3 235B A22B Thinking 2507 or Qwen3.5 122B A10B?
Qwen3 235B A22B Thinking 2507 and Qwen3.5 122B A10B match in llmrun Score, with values of 108.1 versus 109.4; neither fits 24 GB. Choose Qwen3 235B A22B Thinking 2507 for stronger at science and also does chat; choose Qwen3.5 122B A10B for stronger at reasoning and ahead on the benchmarks.
Which needs less VRAM, Qwen3 235B A22B Thinking 2507 or Qwen3.5 122B A10B?
At Q4_K_M, Qwen3 235B A22B Thinking 2507 requires 141.6 GB while Qwen3.5 122B A10B needs 75.4 GB, making Qwen3.5 122B A10B the easier model to run on your local setup.
Can I run Qwen3 235B A22B Thinking 2507 or Qwen3.5 122B A10B on a 16 GB GPU?
Qwen3 235B A22B Thinking 2507 does not fit (smallest quant needs 72.4 GB); Qwen3.5 122B A10B does not fit (smallest quant needs 34.9 GB).
Which has a longer context window, Qwen3 235B A22B Thinking 2507 or Qwen3.5 122B A10B?
Qwen3 235B A22B Thinking 2507 supports 262,144 tokens and Qwen3.5 122B A10B supports 262,144 tokens.